Vegetation, hydrological, and morphological dynamics of natural islands in the Danube Delta

The study was conducted on seven islands of the Danube Delta (Figure 1) in 2023 and 2024. These islands are distributed as follows: one island located (A1) on the Chilia branch with an area of 10.91 ha; three islands located (A2, A3, and A4) in the vicinity of Isaccea port, between Cotul Pisicii and Ceatal Izmail which have an area of 69 ha; and three islands (A5a, A5b, A6) located on the Sf. Gheorghe branch with an area of 32.54 ha.

Figure 1:

Study areas.

Slika 1: Preučevana območja.

The size of the plots was 100 m² (Cristea et al., 2004). The vegetation was classified using the Agglomerative Hierarchical Clustering method (ß-flexible method, β = −0.25 and Bray-Curtis dissimilarity) based on square root transformed data. The initial data were represented by average percentage values corresponding to the Braun-Blanquet (1964) cover-abundance scale (Cristea et al., 2004): r (0.05%), + (0.5%), 1 (5%), 2 (17.5%), 3 (37.5%), 4 (62.5%), and 5 (87.5%).

The dendrogram was made in the GINKGO software (Bouxin, 2005), and the optimal number of clusters was identified based on the average Silhouette index (Rousseeuw, 1987). The synoptic table was obtained from the Juice program version 7.1 (Tichý, 2002).

The coenotaxonomic classification of plant communities is defined according to Chifu et al. (2014a, 2014b), Chifu & Irimia (2014). For each cluster, plant species were presented according to the phi-coefficient value (Chytrý et al. 2002; Tichý & Chytrý 2006). A fidelity threshold of phi = 0.4 and Fisher’s exact test (p < 0.05) were used.

The nomenclature of plant species followed Euro+Med (2024), and the nomenclature of associations followed Chifu et al. (2014a, 2014b), and Chifu & Irimia (2014). The nomenclature of higher syntaxa followed Mucina et al. (2016). Habitat types were classified using the EUNIS habitat classification expert system (Chytrý et al., 2020).

In the study, bathymetric measurements were made using the SonTek RiverSurveyor M9 equipment, mounted on a HydroBoard II. The equipment uses acoustic pulses to determine water velocity, depth, flow, and riverbed elevations based on advanced technology. Multiple acoustic frequencies ensure accurate measurements at shallow and deep depths, even while moving. Data were collected on longitudinal and transverse sections at distances of 100–150 m and in critical areas at 50 m to obtain a detailed representation of the riverbed profile. The equipment features GPS RTK for precise real-time positioning (±3 cm), transmitting and correcting data via Bluetooth. The equipment’s compass was calibrated to compensate for magnetic interference, and the sensors were positioned upstream of the boat’s engine to avoid the influences of vibrations and propeller-generated currents (Sočuvka, 2024). For accuracy, additional topographic measurements were also performed using the Spectra Precision SP80 GPS and the LEICA NA 524 optical level to validate the water level. The collected data provides detailed information on the bathymetric profile, current velocity, and discharge of the riverbed’s cross-sections, ensuring a general picture of the hydrological parameters of the studied area. Violin plots of hydrological factors were produced in R Statistical Software (v4.1.4; R Core Team 2024) via ‘ggplot2’ v.3.5.1 (Wickham, 2016).

The analysis of lower Danube River Island dynamics within this study is conducted through morphometric methods based on high-resolution remote sensing techniques. This approach allows for the precise quantification of changes, whether growth, erosion, or migration, through the detailed comparison of data from various periods. The Danube River islands within this study constitute a complex natural laboratory for the investigation of fluvial morphodynamic processes, reflecting the dynamic interaction between hydrological forces and sediment transport (Leopold & Wolman, 1957; Bondar et al., 1992; Church & Ferguson, 2015). The comparative analysis, utilizing morphometric methods with data derived from ANCPI orthophotoplans (2011 & 2021) and Airbus imagery (2006–2020), reveals a significant variability in their spatio-temporal evolution.

Four EUNIS habitat types are represented in: T11 Temperate Salix and Populus riparian forest; V38 Dry perennial anthropogenic herbaceous vegetation; Q51 Tall-helophyte bed; and Q54 Inland saline or brackish helophyte.

The analyzed vegetation is classified into four phytocoenological classes: Artemisietea vulgaris, Lemnetea, Phragmito-Magnocaricetea, and Salicetea purpureae. In the seven natural islands analyzed, six associations and one sub-associations were identified, including dominant and characteristic species for each habitat.

• Artemisietea vulgaris Lohmeyer et al. in Tx. ex von Rochow 1951

•   Agropyretalia intermedio-repentis T. Müller et Görs 1969

•     Convolvulo arvensis-Agropyrion repentis Görs 1967

•       Convolvulo-Agropyretum repentis Felföldy 1943

• Lemnetea O. de Bolòs et Masclans 1955

•   Lemnetalia minoris O. de Bolòs et Masclans 1955

•     Lemnion minoris O. de Bolòs et Masclans 1955

•       Lemno-Azolletum filiculoides Br.-Bl. 1952

• Phragmito-Magnocaricetea Klika in Klika et Novák 1941

•   Oenanthetalia aquaticae Hejný ex Balátová-Tuláčková et al. 1993

•     Eleocharito palustris-Sagittarion sagittifoliae Passarge 1964

• Bolboschoenetum maritimi Eggler 1933

•   Phragmitetalia Koch 1926

•     Phragmition communis Koch 1926

•       Glycerietum maximae Hueck 1931

• Salicetea purpureae Moor 1958

•   Salicetalia purpureae Moor 1958

•     Salicion albae Soó 1951

•       Salicetum albae Issler 1926

•         amorphoetosum fruticosae Morariu et Danciu 1970

•       Rubo caesii-Salicetum albae Doniță et Dihoru 1961 em. Doniță et al. 1966

The data obtained from the cluster analysis were presented as a dendrogram (Figure 2) and a synoptic table (Table 1). The vegetation was grouped into seven clusters and assigned to different syntax groups described in the literature.

Figure 2:

Dendrogram of vegetation data obtained from cluster analysis.

Slika 2: Dendrogram vegetacijskih podatkov, narejen s klastrsko analizo.

Water level (Figure 3; Table 2). The water level measured at different points showed significant seasonal fluctuations. At Isaccea, levels remained relatively stable in spring and summer (approximately 2.86 m in spring and 2.73 m in summer). However, it increased considerably in autumn and winter, reaching 4.12 m in winter. The same trend was observed at locations such as Dunavăț and Dranov, where winter values were higher than summer values (for example, at Dranov, the water level increased from 0.87 m in summer to 2.00 m in winter).

Figure 3:

Variation of hydrological measurements in the analyzed areas.

Slika 3: Variacija hidroloških meritev na preučevanih območjih.

Water velocity (Figure 3; Table 2). The water velocity had prevailing values between 0.3 and 0.9 m/s throughout the year, suggesting a consistency in moderate flow areas. The highest velocities were recorded in autumn and winter, reflecting the water accumulations in these periods. At Isaccea, the water velocity increased constantly throughout the year, from spring to winter, which indicates an intensification of the water flow, especially during the rainy season and snowmelt. Thus, at Isaccea, the water velocity was 0.68 m/s in spring, decreased slightly in summer to 0.66 m/s, but showed a significant increase in autumn (0.81 m/s) and reached 0.88 m/s in winter. In Dunavăț and Dranov, the water velocity was lower but followed the same trend. At Dranov, the water velocity increased from 0.13 m/s in summer to 0.27 m/s in winter.

Flow measurements (Figure 3; Table 2). At Isaccea, the water flow was 6052 m³/s in spring and decreased slightly in summer to 5876 m³/s. It increased considerably in autumn (7171 m³/s) and reached a maximum of 7859 m³/s in winter. At Dranov, the flow was significantly lower but followed a similar trend, increasing from 508 m³/s in summer to 1089 m³/s in winter.

Riverbed profiles (Figure 3; Table 2). Bathymetric measurements indicate the elevation of the riverbed, helping to assess changes in its structure. At Isaccea, Dunavăț, and Dranov, the values were relatively constant throughout the year, with slight variations, suggesting the stability of the riverbed in these areas. At Isaccea, the depth of the riverbed varied very little from −6.52 m in spring to −6.50 m in winter, indicating sediment accumulation. At Dranov, the values showed a linear trend, ranging from −3.87 m in spring to −3.99 m in winter. The intermediate range remained relatively constant in all seasons, varying very little between −3.87 and −3.99 m.

The Danube Delta, a dynamic fluvial-marine environment, showcases a complex interplay of erosion and accumulation, influencing the morphology of its numerous islands. An analysis of seven islands, reveals distinct patterns of sediment redistribution. Particularly, all islands exhibit a net positive accumulation, a testament to the delta’s robust depositional regime. However, the magnitude of this accumulation varies significantly, highlighting the influence of local hydrodynamics and island positioning.

The recent emergence of A5a and A5b islands, around 2010, along with the substantial accumulation observed in A1, A2, and A3, suggests a modification of this regime, most likely associated with a significant change in the amplitude of the water level and river discharge (Figure 4). Principally, decreases in flow promote rapid sediment deposition. These hydrological events favored rapid sediment accumulation, leading to the formation and expansion of the islands. In this respect, island A5a demonstrated a spectacular area increase, from 3.64 to 18 ha in 10 years, surpassing the expansion rate of island A5b (Figure 4), while A4 demonstrates a dual pattern of erosion and sedimentation. This pronounced accumulation may be attributed to its location within the rectified Perivolovca meander, where altered flow dynamics promote enhanced sediment deposition. In contrast, A2 exhibits the lowest accumulation, only 0.28 ha, concentrated on its right bank, shielded from the direct river discharge. This island, however, experiences 0.88 ha of erosion on its left bank, facing the primary river flow, illustrating a localized disparity in morphometric processes. A1, initially showing exclusive accumulation, has since migrated and become a definitive part of the Danube’s right bank. A2 and A3 display both erosion and accumulation, with accumulation exceeding erosion. This variability highlights the local influences of flow conditions and sediment availability.

Figure 4:

The location of the study’s islands along the lower Danube River – islands morphodynamics.

Slika 4: Lokacija preučevanih otokov ob spodnjem teku reke Donave – morfodinamika otokov.

The tendencies of sedimentation and stabilization are observable in the advancement of islands A5a and A5b toward the river banks, and in the land-tied evolution of islands A4 and A1 to the Danube’s right bank. A5b, also located within the rectified Perivolovca meander, demonstrates a significant accumulation of 7.83 ha, further supporting the fact that meander rectification can amplify sediment deposition. A6, situated in the rectified Dranov 1 meander, displays a similar trend, with 7.69 ha of accumulation and no recorded erosion, reinforcing the influence of altered channel geometry on sediment dynamics.

Although less widespread than accumulation, erosion remains a noteworthy factor. The dual dynamics, characterized by upstream erosion, where river erosion is strongest, and downstream sedimentation, as observed in island A4. Thus, the first half of island A4 was eroded of 1.65 ha in its upstream over 10 years, while the downstream half exhibited emergent zones totaling 3.9 ha within the same period (Figure 4). Conversely, A5a registers minimal erosion, a mere 0.07 ha. A3 and A2 show intermediate erosion values, 0.33 and 0.88 ha respectively, indicating localized erosion patterns influenced by their position relative to the river’s flow. A5b displays 0.26 ha of erosion. A1 and A6 register zero erosion. The varied sedimentation rates, again, evidenced by the spectacular area growth of island A5a, suggest an intense sediment transport dynamic by the river, exceeding the expansion rate of island A5b.

Furthermore, the land-tied evolution observed in A1 and A4 underscores the dynamic nature of island development. A1, exhibiting 6.99 ha of accumulation, is undergoing land-tied evolution towards the right bank, indicating a process of gradual connection to the mainland. Similarly, A4, with 3.91 ha of accumulation in its downstream segment, displays land-tied evolution towards the right bank and underwent significant changes along its left bank at fluctuating rates being exposed to the river’s discharge. These processes are indicative of the delta’s continuous shaping by sedimentary processes. The predominant upstream extension of island A5a indicates progressive sedimentation of the Perivolovca meander, while lateral extension and bank land-tied evolution, observed in island A5b, reflect lateral deposition processes (Figure 4).

In the analysed islands, the communities of Salix alba and Amorpha fruticosa predominate. According to Sitzia et al. (2018), S. alba has an important role in the ecology of the Danube Delta. Thus, through its root structure, S. alba can increase soil resistance and provide additional support through bioengineered structures (Zhang et al., 2013). At the same time, it has been shown that at high phosphorus concentrations (Hrynkiewicz et al., 2015), a decrease in ectomycorrhizal colonization of Salix viminalis and Betula pendula seedlings was recorded. In contrast, Amorpha fruticosa can fix nitrogen and enrich the soil with nitrogen compounds, which causes changes in natural habitat conditions (Gudžinskas & Žalneravičius, 2015). This species also has significant regeneration ability (Avramidou et al., 2023) and is a threat to native ecosystems (Nagy et al., 2018).

Riparian vegetation is strongly influenced by hydrological factors (Dijk et al., 2013; James et al., 2016; Zhu et al., 2017) along river systems. Variation in flow patterns and water depth influences seed distribution, which can affect vegetation stability (Dijk et al., 2013; Zhu et al., 2017). Also, topographic factors and landscape structure control the hydrological response, which will subsequently shape the spatial distribution and diversity of riparian vegetation (Jencso & McGlynn, 2011; Hwang et al., 2012).

In the study, the data obtained showed a significant seasonal fluctuation of the water level, with minimum values in summer and spring, and maximum values in autumn and winter. These values are comparable with the data in the literature (Romanescu, 2014; Trifanov et al., 2019), which show that precipitation and anthropogenic processes influence the water levels in the Danube Delta. In this context, the variations recorded in this study confirm the characteristic hydrological dynamics of the Delta, being highlighted by seasonal water accumulations. Regarding the water velocity, the data from this study showed an increase in the water velocity in autumn and winter. These results are consistent with the observations in the literature (Pekárová et al., 2019; Crețescu et al., 2021) which show that snowmelt and intense precipitation upstream contribute to the variation of natural factors. Moreover, constant and moderate velocities in the other seasons indicate hydrological stability, in which the redistribution of the flow between the Chilia, Sulina, and Sf. Gheorghe arms influences the water flow regime (Pekárová et al., 2019). In the case of water flow, a significant increase was observed in winter.

According to the literature (Pekárová et al., 2019; Crețescu et al., 2021), the Danube acts as a major water collector, in which the discharges are influenced by climatic conditions and flows from upstream countries. At the same time, our study indicates a significant correlation between flow and water velocity during periods of maximum water accumulation. The data obtained for riverbed profiles showed the stability of the riverbed, with minimal variations in depth. According to the literature (Romanescu, 2014; Bănescu et al., 2020), the stability of the riverbed profile is the result of geomorphological processes. In this context, our results confirm that the riverbed is stable and characterized by reduced erosion processes.

The observed morphogenetic processes are closely linked to the Danube’s hydrological regime (Leopold & Maddock, 1953; Bondar, 1992). The intensified dynamics of these sediments are evidenced by considerable changes in shape and area, reflecting their sensitivity to hydrological fluctuations (Knighton, 1998; Gordon et al., 2004). With this sediment accumulation, the rapid colonization of emergent vegetation plays a vital role in soil stabilization, contributing to sediment fixation and the consolidation of island morphology (Bondar, 1992). The dual dynamics, characterized by upstream erosion, reflect a dynamic and unstable morphological equilibrium, influenced by local variations in current and sediment distribution patterns (Panin, 1996). The direction of island A4 expansion is controlled by the complex interaction between water level, velocity, and current direction (Knighton, 1998).

The morphodynamic analysis of the Danube Delta’s islands show the important role of the hydrological regime in shaping this complex ecosystem, highlighting the river’s capacity to induce rapid transformations through sedimentation and erosion processes. The comparative study, utilizing geomorphological methods and remote sensing, has indicated the complex interactions among hydrological, sedimentary, and vegetative factors, offering a detailed perspective on the riverbed including island’s morphological evolution. The data obtained through morphometric analysis signals the necessity for sustainable management, which takes into account the river’s natural dynamics and the potential impact of climate change. Finally, the remarkable intensity of morphodynamic processes, manifested by accelerated rates of sedimentation, erosion, and island shape modification, underscores the highly dynamic nature of the Danube River, where even minor hydrological fluctuations can generate significant morphological transformations within a short timeframe (Bondar et al., 1992; Church & Ferguson, 2015).